[
https://issues.apache.org/jira/browse/HDFS-13281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16418094#comment-16418094
]
Rushabh S Shah commented on HDFS-13281:
---------------------------------------
bq. How does HDFS-12597 use it?
The use case is:
1. EZ-aware webhdfs client (i.e add header:X-Hadoop-Accept-EZ) will send
{{createFile}} request to namenode.
2. If the client supports webhdfs and if the file in EZ, then namenode will
return {{FeInfo}} in response via header and append "/.reserved/raw" to
redirect path.
3. The client will encrypt data with {{FeInfo}} and stream the encrypted bytes
to namenode.
4. Since the path is prepended with {{/.reserved/raw}}, datanode will not
encrypt again.
5. At the end, client will issue {{setXAttr}} on path to namenode.
6. According to HDFS-13035, we will allow owner of the file to do {{setXAttr}}
_only if it is not set_. If namenode will {{setXAttr}} even on
{{/.reserved/raw}} then webhdfs client will fail to {{setXAttr}}.
Hope it makes sense.
> Namenode#createFile should be /.reserved/raw/ aware.
> ----------------------------------------------------
>
> Key: HDFS-13281
> URL: https://issues.apache.org/jira/browse/HDFS-13281
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: encryption
> Affects Versions: 2.8.3
> Reporter: Rushabh S Shah
> Assignee: Rushabh S Shah
> Priority: Critical
> Attachments: HDFS-13281.001.patch
>
>
> If I want to write to /.reserved/raw/<dir> and if that directory happens to
> be in EZ, then namenode *should not* create edek and just copy the raw bytes
> from the source.
> Namenode#startFileInt should be /.reserved/raw/ aware.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
